Controlled benchmark clarifies failures in language-model data attribution

The paper studies why language-model data-attribution methods fail, noting evaluations often vary several dataset properties at once. It introduces CATT, a controlled fact-tracing benchmark using fine-tuning on entirely fictitious knowledge with unambiguous ground truth. Why it matters Reliable attribution is needed to trace model outputs to... Weiterlesen: Controlled benchmark clarifies failures in language-model data attrib…